A retrospective case–control study of gastrostomy use in children undergoing hematopoietic cell transplantation

Maintaining a good nutritional status during the hematopoietic cell transplantation (HCT) procedure is challenging in the pediatric population.


Gastrostomy insertion, either as Percutaneous Endoscopic
Gastrostomy (PEG), or as laparoscopic or minor open surgical procedure, has been widely used in children with a range of chronic disorders such as feeding disturbances, malnutrition, and metabolic diseases with special dietary regimens, and is well documented, also in the pediatric oncology patients. [4][5][6] However, surgical procedures are not without risk, in particular in children and adolescents with malignant diseases. 7 Nevertheless, in experienced hands, gastrostomy procedures can safely be performed with good results and with few complications. 8 In children undergoing HCT, the literature is sparse of its efficacy and risks, and there is often a fear of more severe complications such as perioperative incidents, infections, and delayed wound healing in this patient group. To date, no previous study has evaluated if gastrostomy feeding influences the survival and outcome of HCT in children.
In the present retrospective, multicenter study, including 227 pediatric HCT patients our aim was to compare the weight development and HCT-related complications between patients receiving a gastrostomy (GS) and patients not receiving a gastrostomy (NGS).

| Patients
The study included all children (aged <18 years) treated with HCT at the Swedish HCT Centers at Sahlgrenska University Hospital Clinical background in the two groups of patients with versus without gastrostomy is presented in Table 1.

| Methods
This study was a retrospective case-control study. The study population was divided dichotomously into those who had received a gastrostomy (GS) before the start of HCT (n = 112) versus those without gastrostomy (NGS) prior to HCT (n = 115). Five patients received a gastrostomy post-HCT. These patients were included in the NGS group and were censored at the time of insertion of the gastrostomy ( Figure 2).
Clinical data were collected protocol-based from the medical records before transplantation and at follow-up time points 3, 6, and 12 months after HCT. Collected data included body weight, route of nutrition, stage, and grade of acute GvHD and chronic GvHD, number of infections and survival and relapse rate since last observational period. Furthermore, data on engraftment was recorded.

| Definitions
Febrile episode was defined as body temperature >38°C resulting in treatment with intravenous antibiotics.
Neutrophil engraftment was defined as absolute neutrophil count >0.5 × 10 9 cells/L during three consecutive days.
Classification of acute GvHD (aGvHD) with grade and stage was based on the Glucksberg criteria 9,10 and chronic GvHD (cGvHD) based on the criteria by the National institutes of Health, 11 was referred to the level of affection/signs from skin evaluated by erythematous rash (% of body surface area) and gut involvement evaluated by volume of diarrhea (mL/kg/day). Liver involvement was evaluated by bilirubin level in blood. Grading of aGvHD (I-IV) was based on the conglomeration of the stage of aGVHD in each affected organ system.
Weight was evaluated by standard deviation score (SDS). 12

| Statistical methods
Presentation includes both descriptive statistics and comparison between patients with gastrostomy (GS) and patients without gastrostomy (NGS) by inferential statistics. Categorical variables were summarized with count and percentage and numerical variables with mean (SD) or median (IQR) when appropriate.
Power calculation was not performed, due to the retrospective design of the study.
The comparison of weight between the GS group and the NGS group was based on SDS-reference values from birth to 18 years of age according to the Swedish population-based longitudinal reference values study, defined by Wikland et al. 12 Logistic regression was performed to analyze binary outcomes (GvHD, infections, weight, days to engraftment and parenteral nutrition) at timepoint 3, 6, and 12 months. Disease-free survival (DFS) was estimated using the Kaplan-Meier analysis.
Hazard ratio was calculated by the Cox proportion hazard model.
Data analysis was performed using R version 3.4.4 13 and statistical significance was defined as p < .05.

| Ethics
This study was approved by The Regional Ethical Review Board of Stockholm, Sweden (Approval Nb 2017-1365).

| Patients
The final study population included 227 children. The mean age of the study population was 8. The two groups differed as to donor, stem cell source, and conditioning regimen (Table 3). There were more patients transplanted for a hemoglobinopathy in the GS group 19/112 (17%), than in the NGS group 7/115 (6.1%) (p = .023).  In 10 of the 89 patients in the GS group, the gastrostomy was still in place at 12 months after HCT, due to the fact that their oral intake was not sufficient. These 10 patients were fed with a combination of oral nutrition and nutrition by the G-tube, in order to maintain sufficient caloric intake. Height and weight are measured to standard deviation scores (SDS).

| Surgical complications
Of the 112 patients who received a gastrostomy before the HCT,

| Nutrition
Patients in the NGS group received more frequent parenteral nutrition at timepoint for HCT as compared to patients in the GS group

TA B L E 4
Non-relapse mortality, GvHD, Infection, Relapse, PN, and oral nutrition only at 6 months post-HCT.

| HCT-related outcome parameters
The mean duration of neutrophil engraftment was 18. between the two groups.

The incidence of aGvHD grade III-IV and of gastrointestinal
GvHD was equally distributed between the groups ( Tables 3-5). No patients presented with cGvHD during the study period.

| Outcome
Six

| DISCUSS ION
Maintaining a good nutritional status during the HCT procedure with its complications in terms of mucositis, nausea, infections, and gastrointestinal GvHD is challenging in the pediatric population. The lack of previous studies regarding nutritional support during pediatric HCT and the absence of international guidelines, probably contributes to site-specific nutritional traditions in children undergoing HCT. Sweden has four pediatric HCT centers, in one insertion of a gastrostomy is a standard procedure prior to HCT.
Long-term illnesses such as severe malignancy, use of chemotherapy, repeated infections, and heavy medications are clinical situations when gastrostomy should be discussed to avoid malnutrition. In many pediatric oncology centers, gastrostomy insertion is knowdays a clinical routine. 14 However, there is still a fear of TA B L E 5 Non-relapse mortality, GvHD, Infection, Relapse, PN, and oral nutrition only at 12 months post-HCT.
severe side effects, such as infections and surgical complications.
Using gastrostomy inserted well before HCT is a tradition in some Swedish centers. Anyhow, the use of gastrostomy is not well studied in pediatric transplantation.
Comparative studies between the use of gastrostomy versus non-GS in pediatric HCT are rare. This retrospective study describes weight development and the HCT-related complications in 227 Swedish pediatric patients. In our study, we did not observe any difference in one year post-HCT survival or in the incidence of aGVHD in the group of patients with and without GS. Based on these results we conclude that gastrostomy is a safe method to use for feeding pediatric patients during HCT.
Loss of bodyweight in children undergoing HCT is a wellknown clinical problem [14][15][16] and may have multiple causes. One of the main causes, observed in previous pediatric HCT studies is low energy and protein intake. 15 In our study, children with GS had significantly lower body weight as compared to the NGS group at timepoint of HCT. However, the GS group maintained their body weight better than patients without GS during the first year post-HCT.
Young children often experience difficulties in swallowing oral medications. To facilitate intake of oral medication, a nasogastric tube is an option. However, insertion of a nasogastric tube is often associated with discomfort and may require some type of medical sedation to allow it. 17 Long-term use of nasogastric tube is not only associated with bodily discomfort, 18 but also psychological discomfort. 19 We did not investigate the quality of life reported by patients or parents in the present study, but based on the results from other studies it is plausible to conclude that the insertion of a GS also is associated with a better quality of life for the patients with a reduced use of parenteral nutrition during the post-HCT period.
The US Food and Drug Administration (FDA) has issued a warning about using general anesthetics and sedation drugs in young children. 20 The basis for that warning is an association observed between general anesthetics and sedation drugs, and impaired neurocognitive development. 17

This work was supported by fundings from the Swedish Childhood
Cancer Fund: PR2016-0017, and from ALF Uppsala. The authors wish to thank Dr Johan Arvidson in Uppsala for valuable support with the study.

CO N FLI C T O F I NTER E S T S TATEM ENT
The authors declare no competing financial interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.